From: Ming Lei <ming.lei@redhat.com>
To: Jeffle Xu <jefflexu@linux.alibaba.com>
Cc: axboe@kernel.dk, snitzer@redhat.com, caspar@linux.alibaba.com,
io-uring@vger.kernel.org, linux-block@vger.kernel.org,
joseph.qi@linux.alibaba.com, dm-devel@redhat.com, hch@lst.de
Subject: Re: [dm-devel] [PATCH v3 09/11] dm: support IO polling for bio-based dm device
Date: Tue, 9 Feb 2021 11:11:22 +0800 [thread overview]
Message-ID: <20210209031122.GA63798@T590> (raw)
In-Reply-To: <20210208085243.82367-10-jefflexu@linux.alibaba.com>
On Mon, Feb 08, 2021 at 04:52:41PM +0800, Jeffle Xu wrote:
> DM will iterate and poll all polling hardware queues of all target mq
> devices when polling IO for dm device. To mitigate the race introduced
> by iterating all target hw queues, a per-hw-queue flag is maintained
What is the per-hw-queue flag?
> to indicate whether this polling hw queue currently being polled on or
> not. Every polling hw queue is exclusive to one polling instance, i.e.,
> the polling instance will skip this polling hw queue if this hw queue
> currently is being polled by another polling instance, and start
> polling on the next hw queue.
Not see such skip in dm_poll_one_dev() in which
queue_for_each_poll_hw_ctx() is called directly for polling all POLL
hctxs of the request queue, so can you explain it a bit more about this
skip mechanism?
Even though such skipping is implemented, not sure if good performance
can be reached because hctx poll may be done in ping-pong style
among several CPUs. But blk-mq hctx is supposed to have its cpu affinities.
--
Ming
--
dm-devel mailing list
dm-devel@redhat.com
https://www.redhat.com/mailman/listinfo/dm-devel
WARNING: multiple messages have this Message-ID (diff)
From: Ming Lei <ming.lei@redhat.com>
To: Jeffle Xu <jefflexu@linux.alibaba.com>
Cc: snitzer@redhat.com, axboe@kernel.dk, joseph.qi@linux.alibaba.com,
caspar@linux.alibaba.com, hch@lst.de,
linux-block@vger.kernel.org, dm-devel@redhat.com,
io-uring@vger.kernel.org
Subject: Re: [PATCH v3 09/11] dm: support IO polling for bio-based dm device
Date: Tue, 9 Feb 2021 11:11:22 +0800 [thread overview]
Message-ID: <20210209031122.GA63798@T590> (raw)
In-Reply-To: <20210208085243.82367-10-jefflexu@linux.alibaba.com>
On Mon, Feb 08, 2021 at 04:52:41PM +0800, Jeffle Xu wrote:
> DM will iterate and poll all polling hardware queues of all target mq
> devices when polling IO for dm device. To mitigate the race introduced
> by iterating all target hw queues, a per-hw-queue flag is maintained
What is the per-hw-queue flag?
> to indicate whether this polling hw queue currently being polled on or
> not. Every polling hw queue is exclusive to one polling instance, i.e.,
> the polling instance will skip this polling hw queue if this hw queue
> currently is being polled by another polling instance, and start
> polling on the next hw queue.
Not see such skip in dm_poll_one_dev() in which
queue_for_each_poll_hw_ctx() is called directly for polling all POLL
hctxs of the request queue, so can you explain it a bit more about this
skip mechanism?
Even though such skipping is implemented, not sure if good performance
can be reached because hctx poll may be done in ping-pong style
among several CPUs. But blk-mq hctx is supposed to have its cpu affinities.
--
Ming
next prev parent reply other threads:[~2021-02-09 3:11 UTC|newest]
Thread overview: 54+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-02-08 8:52 [dm-devel] [PATCH v3 00/11] dm: support IO polling Jeffle Xu
2021-02-08 8:52 ` Jeffle Xu
2021-02-08 8:52 ` [dm-devel] [PATCH v3 01/11] block: move definition of blk_qc_t to types.h Jeffle Xu
2021-02-08 8:52 ` Jeffle Xu
2021-02-08 8:52 ` [dm-devel] [PATCH v3 02/11] block: add queue_to_disk() to get gendisk from request_queue Jeffle Xu
2021-02-08 8:52 ` Jeffle Xu
2021-02-08 8:52 ` [dm-devel] [PATCH v3 03/11] block: add poll method to support bio-based IO polling Jeffle Xu
2021-02-08 8:52 ` Jeffle Xu
2021-02-08 8:52 ` [dm-devel] [PATCH v3 04/11] block: add poll_capable " Jeffle Xu
2021-02-08 8:52 ` Jeffle Xu
2021-02-08 8:52 ` [dm-devel] [PATCH v3 05/11] block/mq: extract one helper function polling hw queue Jeffle Xu
2021-02-08 8:52 ` Jeffle Xu
2021-02-08 8:52 ` [dm-devel] [PATCH v3 06/11] block/mq: add iterator for polling hw queues Jeffle Xu
2021-02-08 8:52 ` Jeffle Xu
2021-02-08 8:52 ` [dm-devel] [PATCH v3 07/11] dm: always return BLK_QC_T_NONE for bio-based device Jeffle Xu
2021-02-08 8:52 ` Jeffle Xu
2021-02-08 8:52 ` [dm-devel] [PATCH v3 08/11] dm: fix iterate_device sanity check Jeffle Xu
2021-02-08 8:52 ` Jeffle Xu
2021-02-08 8:52 ` [dm-devel] [PATCH v3 09/11] dm: support IO polling for bio-based dm device Jeffle Xu
2021-02-08 8:52 ` Jeffle Xu
2021-02-09 3:11 ` Ming Lei [this message]
2021-02-09 3:11 ` Ming Lei
2021-02-09 6:13 ` [dm-devel] " JeffleXu
2021-02-09 6:13 ` JeffleXu
2021-02-09 6:22 ` [dm-devel] " JeffleXu
2021-02-09 6:22 ` JeffleXu
2021-02-09 8:07 ` [dm-devel] " Ming Lei
2021-02-09 8:07 ` Ming Lei
2021-02-09 8:46 ` [dm-devel] " JeffleXu
2021-02-09 8:46 ` JeffleXu
2021-02-19 14:17 ` [dm-devel] " Mikulas Patocka
2021-02-19 14:17 ` Mikulas Patocka
2021-02-24 1:42 ` JeffleXu
2021-02-24 1:42 ` JeffleXu
2021-02-26 8:22 ` JeffleXu
2021-02-26 8:22 ` JeffleXu
2021-02-08 8:52 ` [dm-devel] [PATCH v3 10/11] nvme/pci: don't wait for locked polling queue Jeffle Xu
2021-02-08 8:52 ` Jeffle Xu
2021-02-08 8:52 ` [dm-devel] [PATCH v3 11/11] dm: fastpath of bio-based polling Jeffle Xu
2021-02-08 8:52 ` Jeffle Xu
2021-02-19 19:38 ` [dm-devel] " Mikulas Patocka
2021-02-19 19:38 ` Mikulas Patocka
2021-02-26 8:12 ` JeffleXu
2021-02-26 8:12 ` JeffleXu
2021-03-02 19:03 ` Mikulas Patocka
2021-03-02 19:03 ` Mikulas Patocka
2021-03-03 1:55 ` JeffleXu
2021-03-03 1:55 ` JeffleXu
2021-02-17 13:15 ` [dm-devel] [PATCH v3 00/11] dm: support IO polling JeffleXu
2021-02-17 13:15 ` JeffleXu
2021-03-10 20:01 ` [dm-devel] " Mike Snitzer
2021-03-10 20:01 ` Mike Snitzer
2021-03-11 7:07 ` [dm-devel] " JeffleXu
2021-03-11 7:07 ` JeffleXu
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20210209031122.GA63798@T590 \
--to=ming.lei@redhat.com \
--cc=axboe@kernel.dk \
--cc=caspar@linux.alibaba.com \
--cc=dm-devel@redhat.com \
--cc=hch@lst.de \
--cc=io-uring@vger.kernel.org \
--cc=jefflexu@linux.alibaba.com \
--cc=joseph.qi@linux.alibaba.com \
--cc=linux-block@vger.kernel.org \
--cc=snitzer@redhat.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.